This editorial appears in the September issue of the American Journal of Bioethics
by David Magnus, PhD
This issue features a set of articles addressing allocation issues for drugs for COVID-19 that have not been approved by the FDA but are available through either Emergency Use Authorization or Expanded Access Programs (EAP). Webb et al. propose a framework to guide these decisions. Of particular note, the authors point out that a requirement for EAP programs is that they do not interfere with clinical development, including clinical trials, and that the larger the number of participants, the greater the evidence required for an EAP.
Convalescent plasma (CVP) is a promising and plausible candidate as a treatment for COVID-19. CVP has a long history as a treatment for a wide range of diseases and is based on a plausible biological mechanism, namely that neutralizing antibodies circulating in the blood could provide either immunity in the uninfected or treatment for those who are. There is some evidence, mostly from case reports, case series, and retrospective analyses, that suggests that CVP is safe and possibly effective for other viral infections, including other coronavirus infections. There is non-human primate data supporting the development of protective immunity after recovery from acute infection. Taken together, there is ample evidence that CVP is a plausible candidate for trials to determine whether or not CVP is a safe and effective treatment for COVID-19, and if so, when in the course of the disease it is effective. The available clinical evidence and biological intuition particularly point to potential benefits early in the course of disease or even as a prophylactic, but it is not unreasonable to explore its benefits for sicker patients, those who require hospitalization and are critically ill (even if there may be somewhat less hope that there will be benefit in this population).
One would expect, given this background, that a major priority would be the rapid design and execution of large, multi-center randomized clinical trials of CVP for both hospitalized patients and outpatients. But in the U.S., randomized trials of CVP have received less attention, funding, and governmental support than we might have hoped. Instead, the FDA approved a large multicenter EAP on April 7th, administered through the Mayo Clinic, which made CVP available in a non-randomized fashion to over 20,000 hospitalized, critically ill patients at over 2,000 sites. Given the heterogeneity of outcomes for critically ill COVID patients across hospitals, what has been learned from the significant number of patients who have been given CVP through the EAP? Very little. There is perhaps more support for the safety of the intervention. And there is enough evidence so that it remains a plausible target of study. However, it remains unknown whether the intervention provides benefit to the patients it is being provided to. Had the federal dollars (through BARDA) that went to this EAP been spent on a randomized, placebo-controlled trial, we would already know whether the intervention works. In contrast, large multi-center randomized clinical trials of CVP in hospitalized patients have been much more rapidly initiated in a coordinated fashion both in the UK and Canada, and are likely to be able to provide us with a definitive answer regarding whether or not this is an effective intervention.
Why did Mayo (the lead for the CVP EAP) and other institutions choose to focus on an objectively inferior research design that was likely going to be unsuccessful in answering the critical question of whether there was benefit to the intervention? Why did BARDA fund the effort? Why was the EAP program approved by the FDA?
Part of the answer can be found in a growing trend in the U.S. to drop research standards for seriously ill patients when there are no effective treatments available. In the Abigail Alliance case, the parents of a terminally ill patient with head and neck cancer (and the advocacy group they created on her behalf) sued the commissioner of the FDA for the right to access an unapproved drug that was then in clinical trials for colon cancer. They argued that there is a constitutional right to access drugs after Phase I. The court eventually found that there is no right to unproven, unapproved drugs, but the battle over access has continued. The FDA has made it easier to access unproven drugs through EAPs, and most states have adopted “right to try laws,” that make it easier to bypass oversight. On May 30, 2018, a federal right to try law was also adopted, allowing patients access to unproven drugs without FDA or IRB review. Pharmaceutical companies have been forced to serve as gate keepers to prevent unwarranted early access to drugs (at least in cases where it is likely to hamper the research needed to obtain full FDA approval of the drug). The case of Josh Hardy, whose family launched a successful campaign to gain access to an experimental drug (eventually leading to the firing of the CEO), and many other similar cases demonstrate how challenging it is for companies to say no to desperate patients and their families who can successfully muster social media to support their efforts.
We have seen this playing out with treatments for COVID 19. Hydroxychloroquine was touted as a treatment for the disease without adequate evidence leading to countless off-label prescriptions. These off-label uses were a potential barrier to recruitment to randomized, placebo-controlled trials that would actually establish whether the drug is safe and effective (and we now have some evidence to suggest that it is neither).
In all of these cases, physicians served as potential gate keepers, and yet they advocated for expanded access, compassionate use exceptions, or off-label prescriptions instead of clinical trials. These physicians presumably knew that the drugs were unproven. And they likely understood the need for randomized trials to establish whether the drugs worked. And they probably understood that their activities contributed to undermining these badly needed trials, whether by undermining trial recruitment (hydroxychloroquine), diverting resources from clinical trials to compassionate use or EAP (the Hardy case), or, as in the case of CVP for COVID-19, both.
The ethos of medicine in the U.S. reflects the underlying ethos of our society. And that ethos is grounded in individualism. People pursue their own interests and rights are largely about prevention of others from infringing on individual liberty. This has led to decades of debate and tension between public health measures necessary for the health of the community versus individual liberties. It plays out over mandates for childhood vaccination, seatbelt laws, and quarantine for certain infections. It is now being fought over mandates for social distancing and mask requirements. Public health experts are puzzled by public opposition to mandates that ask for little sacrifice and promote the greater good. They shouldn’t be surprised.
In medicine, this ethos means that patients will by and large pursue their own interests (or those of their family members) and physicians largely see their role as promoting the welfare of the individual patient in front of them.
From this perspective, it is perhaps understandable that patients or their loved ones would do all in their power to obtain possibly effective treatments when they are critically or seriously ill and there are no interventions with proven efficacy. Even if an intervention is unproven, patients in desperate straits may feel that they have little to lose. Hence the growing chorus to end the FDA approval process after Phase I for potentially life-threatening conditions without treatment.
Similarly, an argument can be made that physicians should advocate for these interventions for their patients who have little to lose. The impact on others is absent from the clinical gaze. If their job is to focus on the welfare of their individual patient, physicians may apply the same calculous to the patient that their loved ones use: they have little to lose. And that same ethos drove the FDA, Mayo Clinic, and BARDA to offer something, anything, to desperate patients who have little to lose.
But the collective consequences of individual decisions have serious repercussions. As in any problem of the commons, individuals acting in their own self-interest can result in everyone being worse off. Failure to acknowledge the communal good as an important part of the responsibility of physicians has resulted in a threat to the integrity of our research enterprise. Without rigorous clinical trials, we will be left not knowing if treatments work or outweigh the risks. Furthermore, even when considering only the benefit/harm ratio for an individual patient, it is important for the physician and patient to consider that CVP has potential risks. Known risks of plasma transfusion include TACO (transfusion associated circulatory overload) and TRALI (transfusion related acute lung injury); there may be additional, unexpected risks specific to CVP. While we await the results of randomized controlled trials, the vast majority of CVP recipients in the US are being transfused according to the EAP, potentially being subjected to harms of transfusion without receiving any benefit. If we find out that there is no benefit to CVP, then we will have subjected tens of thousands of patients to the risks of CVP without providing any benefit.
It is further unclear that an individually focused ethos can adequately address issues of social inequality. The ability to muster social media to support requests for compassionate use is not equally distributed. Similarly, patients will differentially have access to physician advocates and to health provider institutions (such as academic medical centers) that have access to interventions offered through EAP’s.
Society would be well served by a shift in medicine to a more balanced ethos that values communal welfare and solidarity in addition to individual liberty. COVID-19 is a test to determine if the values of society are up to the challenge of a crisis that requires collective action and a focus on the collective good. And we are failing that test.